Spectro-Temporal Post-Filtering Via Short-Time Target Cancellation for Directional Speech Enhancement in a Dual-Microphone Hearing AID

Abstract

The approach to directional speech enhancement described and evaluated herein involves combining adaptive beamforming and spatial spectro-temporal post-filtering using a small dual-microphone end-fire array, as is used in modern digital hearing aids. An established and commonly used hearing aid algorithm is used for the adaptive beamforming while our Short-Time Target Cancellation (STTC) processing, which leverages the computational efficiency of the Short-Time Fourier Transform (STFT), is used for the post-filtering. Our STTC processing is notable for effecting real-time low-latency (≤20 ms) spatial spectro-temporal filtering with causal, efficient and minimalist STFT-based processing that has low requirements in terms of energy consumption, memory and computational power. The evaluation is carried out in both semi-anechoic and reverberant rooms using instrumental measures of speech enhancement. As demonstrated in this brief study, our STTC spatial spectro-temporal filtering can be implemented with hearing aid microphones and can be used to enhance established hearing aid processing.